Which scenario best describes a multilabel classification?

• A. An image contains both a dog and a cat label.
• B. An image contains only one object label.
• C. Data with a single label per instance.
• D. A regression problem predicting weight.

Answer: (A)

Multilabel classification means an instance can belong to several categories at the same time. The scenario where an image is labeled with both a dog and a cat shows multiple labels applying to the same image, which is the core feature of this type of task. In contrast, single-label (or multiclass) classification assigns only one label to each instance, even if multiple objects are present. A regression problem like predicting weight isn’t about assigning discrete labels to instances at all, so it lies outside classification. In multilabel tasks, you typically predict a set of labels (often as a vector of probabilities or yes/no for each possible label) for each instance.